MOLECULAR CARCINOGENESIS 6:s-9 (1992)

BRIEF COMMUNICATION

Differential Effect of Tumor Necrosis Factor on Proliferation of Primary Human Keratinocytes and Cell Lines Containing Human Papillomavirus Types 16 and 18 Luisa Lina Villa,' Katia Brito Lins Vieira, Xu-Fang Pei, and Richard Schlegel Ludwig Institute for Cancer Research, SJo Paulo, Brazil (LLV; KBLV); and Department of Pathologx Georgetown University Medml School, Washington, DC (X-Fe RS) Keratinocytes immortalized by human papillomaviruses (HPV) 16 and 18 are partially resistant t o t h e inhibition of proliferation exerted by transforming growth factor-p (TGF-p). To determine if this finding reflects a generalized resistance t o inhibitory cytokines, w e studied t h e effect o f t u m o r necrosis factor-a (TNF-a) on subconfluent cultures o f b o t h normal and HPV-immortalized human foreskin keratinocytes. Whereas primary and HPV-16-immortalized keratinocytes were sensitive t o TNF-a, HPV-18-immortalized keratinocytes (and those immortalized by simian virus 40) were resistant t o the inhibitory effects o f this cytokine. The ability of HPV-18 to induce a more resistant phenotype correlated with i t s more potent in vitro transforming activity and its appare n t association with more aggressive tumors. Interestingly, t h e state of TNF-induced growth inhibition in normal o r HPV-16-immortalized keratinocytes was n o t accompanied by a reduction in t h e expression o f c-myc RNA or protein. This contrasts sharply with t h e ability of TGF-p to inhibit c-rnyc RNA expression in normal cells. Evidently, t h e resistance o f HPV-immortalized keratinocytes t o TNF-a and TGF-p proceeds along different regulatory pathways. Q 1992 Wiley-Liss. inc. Key words: Papillomavirus, growth inhibition, oncogene expression, c-myc RNA and protein INTRODUCTION

The proliferation and differentiation of human keratinoc.ytes can be modulated in vitro by a variety of positive and negative regulators. Transforming growth factor-p (TGF-p) m d tumor necrosis factor-a (TNF-cY) are two potent inhibitors of keratinocyte proliferation that may have an important role in the in vivo control of keratinocyte growth. Actually, TGF-p is a potent growth inhibitor for most epithelial cells [ I ] and has been shown to selectively reduce t-myc expression through the inhibition of transcriptional initiation involving a cis regulatory element [2,3]. This has led to the hypothesis that the inhibition of keratinocyte qrowth by TGF-p is mediated by the downregulation of c-myc expression, since constitutive expression of c-myc appears to be necessary for keratinocytc proliferation [3]. TNF-a is a 17-kDa protein synthesized by stimulated macrophages and can induce hemorrhagic necrosis of certain mouse and human tumors (41. It is cytotoxic or cytostatic for a variety of cells in vitro, including normal fibroblasts [5], endothelial cells [ 6 ] ,leukemic cell lines [ 7 ] ,and both normal [8] and tumor-derived [9] epithelial cells. Like TGF-(3 and interferon, the action of T N F - a requires specific binding to high-affinity receptors that are expressed on cells of most tissues as well as on their malignant derivatives [lo]. On normal keratinocytes, this binding causes a reversible inhibition of proliferation accompanied by the stimulation of differentiation 181. The mechanism by which TNF-u leads to growth arrest or differentiation of certain cells or both is still poorly understood. The human papillomaviruses (HPVs) are associated with CI 1992 WILEY-LISS, INC

a variety of neoplasms, the most clinically significant being cervical carcinoma [ 1 1,121 HPV DNA is present and actively transcribed in the majority of cancers of the uterine cervix as well as in premalignant lesions of the genital tract [13,14] The "malignancy-associated" HPV types (types 16 and 18) can immortalize normal cells in culture [15,16j, and it has recently been established that two HPV genes E6 and E7, are essential for the efficient immortalization of human cells [I 7-21] The E6 and E7 proteins of these viruses bind to the products of the retinoblastoma [221 and p53 [ 2 3 ] tumor suppressor genes, respectively The cellular phenotype that results from E6iE7 expression varies and depends upon the specific type of HPV DNA used and the method of selecting for the transformed or immortalized cells For example, when selected by their inability to respond to signals for terminal differentiation, human foreskin keratinocytes immortalized with HPV-16, HPV-18, or silriar uirus 40 (SVClO! are resislar'++Cthe growthinhibitory effects of TGF-p In addition, these differentiationresistant cells do not downregulate c-myc transcription in response to TGF-p [24] Genetic studies have demonstrated that the HPV E7 oncoprotein is responsible for this phenomenon, presumably via its interaction with the tumor suppressor protein pRB 1241 In contrast, when HPV'Corresponding author. Ludwig institute for Cancer Research, R. Prof. Antonio Prudente 109-4". 01 509 Sao Paulo-SP, Brazil. Abbreviations DMEM, Dulbecco's modified Eagle's medium, HPV, human papillomavirus, PBS, phosphate-buffered saline, SDS, sodium dodpcyl sulfate; SSC, standard saline citrate, N40, simian virus 40, TGF, transforming growth factor; TNF, tumor necrosis facror

6

VILLA ETAL.

immortalized cells are selected by continued passaging, their response to TGF-p is very similar to that of primary keratinocytes [25,261. It has also been reportedthat nontumorigenic and tumorigenic HPV-16-positive cells differ in their respectiveresponses to TGF-p. Nontumorigenic cells are sensitive, whereas tumorigenic cells (such as the SiHa and Caski cell lines) are resistant [25]. Furthermore, TGF-p downregulates c-myc transcription in all TGF-sensitive cell lines. TNF-a, like TGF-p, inhibits normal keratinocytes, but its effect on HPV-immortalizedcells has not been investigated. In the study presented here, we investigated the growthinhibitory response of normal and HPV-immortalizedhuman foreskin keratinocytes to human recombinant TNF-a. NormaI and HPV- 16-t ransformed celIs were inhibited by TNF-a, whereas HPV-18- and SV40-immortalized cells were resistant. The growth inhibition observed in normal and HPV16-transformed cells was not accompanied by a decrease in the expression of c-myc either at the RNA or protein levels. MATERIALS AND METHODS Cell Culture Primarycultures of human epidermal keratinocytes(NHF) were isolated from newborn foreskin by trypsin treatment as previously described [27]. These cells were maintained in serum-free medium (KGM; Clonetics Corp., San Diego, CA) containing epidermal growth factor and bovine pituitary extract, and were passaged two to six times. Human keratinocyte cell lines immortalized by SV40 (HF-SV40), HPV-16(HF-698 and HF-699). and HPV-18(HF-18 and HF-18 Nco) were grown in a mixture of three parts KGM and 1 part Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% fetal calf serum. These HPV-immortalized cell lines are independent, clonal isolates and were obtained by transfection of NHF derived from different foreskins. HeLa cells were cultivated in DMEM containing 10% fetal calf serum. Biological Assay Cells were plated in 60-mm dishes (about 4 x l o 4 cells/dish) and grown until 40-60% confluent, when different concentrations (1 -5 nM) of human recombinant TNF-a (Boehringer Mannheim Corp., Indianapolis, IN) were added. After 2 d, the cells were labeled overnight with 10 p,Ci/mL [methyl-3H]thymidine (25 Ci/mmol; Amersham Corp., Arlington Heights, IL). The cultures were then rinsed with phosphate-buffered saline (PBS) and fixed for 10 min with cold ethanol. After another wash with PBS, the cultures were solubilized with 0.1 N NaOH, neutralized with an equivalent volume of 0.1 N HCI, and assayed by liquid scintillation for incorporation of radioactivity. Northern Hybridization Analysis Cells were grown in 100-mm dishes to 40-60% confluency. After treatment with TNF-a for 4, 8, 20, or 40 h, total RNA was extracted by a guanidinium-CsCI procedure, electrophoresed in hydroxyrnethylmercury gels, and trans-

ferred to nylon membranes (Amersham Corp.). The resultant blots were hybridized overnight with 32P-labeledc-myc DNA (kindly provided by Dr. Nancy Hynes, Friedriesh-Miescher Institute, Basel, Switzerland) or with 32P-labeledp-actin DNA at 42°C in 50% formamide, 2 x standard saline citrate(SSC; 300 m M sodium chloride and 30 m M sodium citrate), 5 x Denhardt's solution (0.1YO Ficoll, 0.1% polyvinylpyrrolidone, and 0.1 YO bovine serum albumin), 0.1% sodium dodecyl sulfate (SDS), and 100 pg/mL denatured salmon sperm DNA. Radioactive probes were prepared with a random-primer DNA labeling system (GIBCOIBRL, Gaithersburg, MD) and used at specificactivitieshigherthan 1 x 109cpm/pg. Membranes were washed twice a t room temperature in 2 x SSC and 0.1YO SDS for 15 min each time and twice at 68°C for 20 min each time in 0.1 x SSC and 0.1 YO SDS. Protein Extraction and Electrophoresis For protein extraction, cells were lysed directly on culture dishes using l .O mL of cold lysis buffer (50 m M HEPES, pH 7.0, 250 m M NaCI, 0.1 YO Nonidet P40), 1'/o phenylmethylsulfonyl fluoride, and 1 rngImL aprotinin and leupeptin for 30 min on ice. Lysates were clarified by centrifugation in an Eppendorf centrifuge for 10 min at 4"C, and 200-pg aliquots were separated by SDS-gel electrophoresis. The SDS gels were then electroblotted onto nitrocellulose membranes, which were reacted with monoclonal antibody specific for the human c-myc protein [28] (from hybridoma cell line MYC CT 14-G4.3 (American Type Culture Collection CRL 1727)). Membranes were developed with the PROTOBLOTwestern-blot alkaline-phosphatasesystem (Promega Biotech, Madison, WI) according to the manufacturer's instructions. RESULTS Effect of TNF-a on Keratinocyte Proliferation To test the effect of TNF-a on keratinocyte growth, subconfluent cultures of normal and HPV-immortalized human foreskin keratinocytes were exposed to TNF-a for 2 d. DNA synthesis in normal primary keratinocytes was inhibited by about 60% after treatment with 5 nM TNF-a (Figure 1). Rapidly growing HPV-16-immortalized cells (HF-698) were even more sensitive to TNF exposure. At a concentration of 1 nM, TNF inhibited the proliferation of HF-698 by more than 60%. In contrast, human keratinocytes immortalized by either HPV-18 (HF-18 Nco and HF-18) or 9/40 (HF-SV40) were more resistant to the cytotoxic effect of the same concentrations of TNF-a. HF-18 Nco cells were only inhibited by about 30%, whereas HF-18and HF-9/40 were even more resistant. To determine whether the response of HF-698 to TNF-a could be due to a particular characteristic of this immortalized cell clone, a second independent clone of HPV-16-transformed cells (HF-699)was also analyzed, and exhibited virtually the same inhibition by TNF-a as HF-698 (data not shown). Effect of TNF-a on c-mycTranscription It has been postulated that the proliferation of human keratinocytes is regulated by the expression of the c-myc

EFFECTOF TNF ON HPV-IMMORTALIZED CELLS

7

Figure 1. Effect of TNF-a on keratinocyte proliferation. Normal and HPV-16-, HPV-18-, and N40-transfected keratinocytes were exposed to two concentrations of TNF-a for 2 d, and then

cell proliferation was measuredby incorporationof [3H]thymidine. The bars represent the mean of three experiments for each condition. The error bars indicate the range of variation.

gene, since myc antisense oligonucleotides inhibit keratinocyte growth [3].To assess whether the antiproliferative effects of TNF-a were accompanied by alterations in c-myc RNA levels, total cellular RNA was prepared from cells exposed to 1 nM TNF-a and analyzed by northern blotting using a myc cDNA probe. As shown in Figure 2, the constitutive level of myc RNA did not change in either the TNF-sensitive normal human foreskin keratinocytes (NHF) or HF-698 cells after exposure to TNF-a. As anticipated, there was also no inhibition of c-myc RNA expression in the HF-SV40 or HF-18 Nco cell lines.

DISCUSSION

c-myc Expression in the Presence of TNF-a Since no alteration in c-myc RNA was observed in the presence of TNF-a, we further investigated the effect of this cytokine on the c-myc protein levels. Protein extracts were prepared at the same time intervals used for the RNA studies and analyzed on western blots using a monoclonal antibody reactive with human myc protein. Figure 3 shows the two characteristic protein products that result from differential splicing between exon 1 and 2 of the c-mycgene. The prominent upper band of myc protein is indicated by the arrow in the right margin of the figure. For all cell lines and conditions examined, no significant alteration of the myc levels was detected. Although the amount of rnyc protein observed in cell extracts was low, these results were consistent and reproducible. In addition, the identity of the protein bands labeled "Myc" in Figure 3 was confirmed using extracts of HL-60 cells, which overexpress this protein (data not shown).

To investigate whether HPV infection of epithelial cells is associated with resistance to additional negative regulators of keratinocyte growth, we compared the response of both normal and HPV-immortalized foreskin keratinocytes to TNF-a. We were able to show that HPV-16-transfected cells were as sensitive as normal keratinocytes to TNF-a growth inhibition but that HPV-18- or SV40-immortalized cells were resistant. Since continuous expression of myc seems to be a prerequisite for epithelial cell growth, we looked at the c-myc levels in cells exposed to TNF-a. No differences could be detected among either sensitive (NHF and HF-698) or resistant cells (HF-18 Nco, HF-18, and HF-SV40) (Figure 2). This result indicated that TNF-a did not downregulate c-mycexpression IikeTGF-p does. However, myc protein might be regulated by posttranscriptional mechanisms. We therefore determined the myc protein levels by immunoblotting using an anti-myc monoclonal antibody. Figure 3 shows that the myc protein levels are the same irrespective of the cell line or amount of TNF-a used. It seems, therefore, that TNF-a does not inhibit proliferation in these cells through alterations in myc protein levels. Expression of the c-myc proto-oncogene has been found to be closely regulated and coupled to the control of growth and differentiation in a variety of cell types. Interestingly, however, c-mycexpression remains unchanged under growing conditions or induction of terminal differentiation in normal mouse primary keratinocytes [29].Whereas keratinocyte differentiation is not accompanied by the down-

8

VILLA ETAL.

Figure 2 . Effect of TNF-u on c-myc mRNA levels in keratinocytes. Total RNAwas extracted from cells exposed to 1 nM TNF-a for the periods of time (in h) indicated at the top of the gels. Ten micrograms of total RNA was loaded in each lane and ana-

lyzed by comparing c-mycsignals with actin mRNA levels. Ribosomal RNA markers are indicated at the right. The numbers a t the bottom of each lane are the intensities of the c-mycsignals relative to the actin signals.

regulation of myc expression, keratinocyte growth inhibition by TGF-P is. This contrasts with the induction of both c-fos and c-jun RNA during growth inhibition by TGF-p [ 2 5 ] . The basis for the difference in resistance of HPV-16- and HPV-18-immortalized keratinocytes t o TNF-a is n o t yet

ble for the greater in vitro transforming activity of HPV-18 [30,31]compared with HPV-16. The results presented in this paper indicate that the TNF-resistant phenotype may reflect the strength of the HPV promoter. Interestingly, 9/40 (which encodes a promoter that is ten times more active

known. However, recent studies have determined that the noncoding region of HPV-18 is predominantly responsi-

than HPV-18) produced the same resistant phenotype. We are presently analyzing additional HPV-immortalized cell

Figure 3. Effect of TNF-u on myc protein levels in different keratinocyte lines. Protein was extracted from cells exposed t o 1 and 5 nM TNF-u for 2 d and analyzed by western blotting using a monoclonal antibody reactive with human c-myc. Each lane

contains 170 pg of total protein extract. Molecular-sizestandards are indicated in the left margin; the position of the myc band is indicated by "Myc" in the right margin.

EFFECT OF TNF ON HPV-IMMORTALIZED CELLS

lines containing smaller portions of the HPV genome and are obtaining similar results. It is possiblethat the 10-fold greater transcriptional activity of the HPV-18 promoter [321 could effect higher levels of the HPV oncoproteins and thereby alter the cellular phenotype, including resistance to inhibitory cytokines. Interestingly, HPV-I 8-containing cervical carcinomas appear to exhibit more aggressive clinical behavior [33,34]. Whether the aggressiveness of cervical tumors is related to their resistance to inhibitory cytokines is unknown, but resistance to TNF-a might contribute to the development of the neoplastic phenotype. In conclusion, we report that TNF-a inhibited the growth of normal human keratinocytesin a way that did not appear to involve downregulation of c-myc expression. In addition, the cell phenotypes induced by the malignancyassociated HPV-16 and HPV-18 viruses differed with respect to TNF-a sensitivity. Further studies will be needed to determine the complex mechanisms of keratinocyte growth control exerted by various classes of inhibitory cytokines. ACKNOWLEDGMENTS

K.B.L.V. is a recipient of a fellowship from Conselho Nacional Para o Desenvolvimento Cientifico e Tecnologico (CNPq). This study was supported in part by Grant 1 R 0 1 CA53371-02 from the National Cancer Institute. Received kbruaty28, 1992, revised March 22, 1992, accepted March 24, 1992

REFERENCES 1 Lyons RM, Moses HL Transforming growth factors and the regulation of cell proliferation Eur J Biochem 187 467-473, 1989 2 Coffey RJ Jr, Bascom CC, Alpes NJ, Graves-Deal R, Weissman BE, Moses HL Selective inhibition of growth-related gene expression in murine keratinocytes by transforming growth factor beta Mol Cell Biol8 3088-3093, 1988 3 Pietenpol JA, Holt JT, Stein RW, Moses HL Transforming growth factor beta1 suppression of c-mycgene transcription Role in inhibition of keratinocvte Droliferation Proc Natl Acad Sci USA 87 3758-3762,1940 ' 4 Old U Tumor necrosis factor (TNF) Science 230 630-632, 1985 5 Tsulimoto M, Yip YK, Vilcek J Tumor necrosis factor Specific binding.and internalization in sensitive and resistant cells Proc Natl Acad Sci USA82:7626-7630, 1985. 6. Robaye B, Mosselmans R, Fiers W, Dumont JE, Galand P Tumor necrosis factor induces apoptosis (programmed cell death) in normal endothelial cells in vitro. Am 1 Pathol 138.447-453, 1991 7. Kronke M. Schluter C, Pfizenmaier K. Tumor necrosisfactor inhibits MYC expression in HL-60 cells at the level of mRNA transcription. Proc Natl Acad Sci USA 84.469-473, 1987. 8. Pillai S,Bikle DD, Eessalu TE, Aggarwal BB, Elias PM. Binding and biological effects of tumor necrosis factor alpha on cultured human neonatal foreskin keratinocytes. J Clin Invest 83:816-821, 1989. 9. Yarden A, Kimchi A. Tumor necrosis factor reduces c-mycexpression and cooperates with interferon-gamma in HeLa cells. Science 234 1419-1421, 1986. 10. Loetscher H, Steinmetz M, Lesslauer W Tumor necrosis factor: Receptors and inhibitors. Cancer Cells 3:221-226, 1991. 11. zur Hausen H. Papillomaviruses in anogenital cancer as a model to understand the role of viruses in human cancer. Cancer Res 49:4677-4681. 1989. 12 Howley PM, Schlegel R. The human papillomaviruses. An overview. Am J Med 85 (suppl 2A): 155-1 58, 1988. 13. Yee C, Krishnan-Hewlett I, Baker CC, Schlegel R, Howley PM. Presence and expression of human papillomavirussequences in human cervical carcinoma cell lines. A m J Pathol 119:361-366, 1986.

9

14. Lehn H, Villa LL, Marziona F, Hilgarth M, Hillemans HG, Sauer G. Physical state and biologic activity of human papillomavirus genomes in precancerous lesions of the female genital tract. J GenVirol69:187-196, 1988. 15 Pirisi L, Yasumoto A, feller M, Doniger J, DiPaolo JA Transformation of human fibroblasts and keratinocytes with human papillomavirustype 16DNA.JViro~61:1061-1066, 1987. 16. Durst M, Dzarlieva-Petrusevska RT, Boukamp P, Fusenig NE, Gissmann L. Molecular and cytogenetic analysis of immortalized human primary keratinocytes obtained after transfection with human papillomavirus type 16 DNA. Oncogene 1.251-256, 1987. 17. Munger K, Phelps WC, Bubb V, Howley PM, Schlegel R. The € 6 and E7 genes of the human papillomavirus type 16 together are necessary and sufficient for transformation of primary human keratinocytes. 1 Virol63:4417-4421, 1989. 18. Hawley-Nelson F: Vousden KH, Hubbert NL, Lowy DR, Schiller IT. HPV-16 € 6 and E7 proteins cooperate to immortalize human foreskin keratinocytes. EMBO J 8.3905-3910, 1989. 19. Barbosa M, Schlegel R. The E6 and €7 genes of HPV-18 are sufficient for inducing two-stage in-vitro transformation of human keratinocytes Oncogene 4: 1529-1 532, 1989. 20. Kaur C McDougall JK, Cone R. Immortalization of primary human epithelial cells by cloned celvical carcinoma DNA containing human papillomavirus type 16 E6iE7 open reading frames. J Gen Virol 70: 1261-1266, 1989. 21. Hudson JB, Bedell MA, McCance DJ, Laimins LA. Immortalization and altered differentiation of human keratinocytes in-vitro by the E6 and €7 open reading frames of human papillomavirus type 18. J Virol 64: 51 9-526, 1990 22. Dyson N, Howley PM, Munger K, Harlow E The human papillomavirus-16 E7 oncoprotein is able to bind to the retinoblastoma gene product. Science 243.934-937, 1989. 23. Werness BA, Levine AJ, Howley PM. Association of human papillomavirus types 16 and 18 E6 proteins with p53. Science 248:76-79, 1990. 24. Pietenpol JA, Stein RW, Moran E, et al. TGF-p1 inhibition of c-myc transcription and growth in keratinocytes is abrogated by viral transforming proteins with pRB binding domains. Cell 61 :777-785, 1990. 25. Braun L, Durst M, Mikumo R, Gruppuso ?i Differential response of nontumorigenic and tumorigenic human papillomavirus type 16-positiveepithelialcellstotransforminggrowthfactor~1.Cancer Res 50:7324-7332, 1990. 26. Woodworth CD, Notario V, DiPaolo JA. Transforming growth factors beta 1 and 2 transcriptionally regulate human papillomavirus (HPV) type 16 early gene expression in HPV-immortalized human genital epithelial cells. J Virol 64:4767-4775, 1990. 27. Schlegel R, Phelps WC, Zhang YL, Barbosa MS Quantitative keratinocyte assay detects t w o biological activities of human papillomavirus DNA and identifies viral types associated with cervicalcarcinoma. EMBOJ 7:3181-3187. 1988. 28. Evans GT, Lewis GK, Ramsay G, Bishop JM. Isolation of monoclonal antibodies specific for tumor c-myc protooncogene product. Mol Cell Biol 5:3610-3615, 1985. 29. Dott GF: Gilman MZ, Maruyama M, Weinberg RA c-mycand c-fos expression in differentiating mouse primary keratinocytes. EMBO J 5.2853-2857.1986. 30. Villa LL, Schlegei R. Differences in transformation activity between HPV-18 and HPV-16 map to the viral LCR-E6-E7 region. Virology 181:374-377, 1991. 31. Rornanczuk H, Villa LL, Schlesel R. Howlev PM. The viral transcriptional regulatory region upstream of the E6 and E7 genes is a major determinant of the differential immortalization activities of human papillomavirustypes 16and 18. JVirol65:2739-2744, 1991. 32. Romanczuk H, Thierry F, Howley PM. Mutational analysis of CIS elements involved in E2 modulation of human papillomavirus type 16 Pg7 and type 18 PIOS promoters J Virol 64:2849-2859. 1990. 33. Barnes W, Delgado G, Kurman RJ, et al. Possible prognostic significance of human papillomavirus type in cewical cancer. Gynecol Oncol29:267-273, 1988. 34. Kurman RJ, Schiffman MH, Lancaster WD, et al. Analysis of individual human papillomavirus types in celvical neoplasia. A possible role for type 18 in rapid progression. Am J Obstet Gynecol 159:293-296, 1988.